Associate Research Scientist

The Ubiquitous Knowledge Processing (UKP) Lab at the Department of Computer Science, Technische Universität (TU) Darmstadt, Germany has several ERC-, HORIZON- or DFG-funded openings for an

Associate Research Scientist (PostDoc- or PhD-level; for an initial term of two to three years)

As an interdisciplinary team, we are happy to accommodate a wide variety of research questions and backgrounds, ranging from machine learning to dataset creation. We particularly welcome proposals on neuro-symbolic reasoning, multilingual and multimodal language modeling, and creative ideas on modeling cross-document relations. We are looking for outstanding PhD and postdoc candidates with a strong profile in Machine Learning and/or Natural Language Processing. Prior relevant research experience and strong engineering skills are a plus. We highly appreciate demonstrable engagement in open-source projects, communication skills in English and the ability to effectively cooperate in an interdisciplinary setting.

The positions are funded by the ERC Advanced Grant “InterText – Modeling Text as a Living Object in a Cross-Document Context”, by the recently approved HORIZON project “Socially-acceptable Extended Reality Models and Systems” (SERMAS) and by the project “UKP-SQuARE: Software for Question Answering Research” funded by the German Research Foundation:

• InterText deals with research questions related to modeling cross-document relations in long documents, including novel problems and task definitions. Machine learning research includes novel neural architectures for processing long documents, injecting documents structure and commonsense knowledge into transformers, and domain adaptation.
• The overarching goal of SERMAS is to develop next-generation extended reality (XR) systems that are socially acceptable (comfortable use, transparency, safety and trustworthy). We develop a multimodal dialogue agent capable of handling natural language inputs and non-verbal information. This work will allow for fluent and natural conversations between users and the agent.
• Finally, UKP-SQUARE is a flexible, scalable and interpretable QA platform to enable researchers to share their custom QA agents by integrating them to our platform using easy-to-use common interfaces, study the strengths and weaknesses of existing models by comparing them on a wide range of tasks and datasets that are already provided within our framework, and explore the existing models and datasets to answer more specific research questions using integrated interpretability tools.

The UKP lab comprises over thirty team members working on cutting-edge NLP and Machine Learning, for example SBERT or AdapterHub. Their novel applications extend to mining scientific literature and social media, applied cybersecurity, or AI for the Social Good in general. The research environment is highly agile, diverse and supportive. There exists a wide cooperation network of both leading academic and industrial partners. The Department of Computer Science of the TU Darmstadt is regularly ranked among the top ones in Germany and offers a premier scientific environment, such as the Hessian Center for Artifical Intelligence hessen.AI, the National Center for Applied Cybersecurity ATHENE, the European Network of Excellence ELLIS und numerous high-profile international and national collaborations including industry.

UKP Lab is committed to cutting-edge research, publishing in top-tier venues, cooperative work style and close interaction of all team members. The selected candidates enjoy diverse opportunities for professional growth, leading both to successful faculty careers and exciting industrial research jobs. Darmstadt is very close to the Frankfurt airport and provides an excellent living environment for incoming staff members. There is an ELLIS unit in Darmstadt, and Iryna Gurevych is co-director of the ELLIS NLP program.

The position assumes that you can move to Darmstadt or a nearby German location. It is currently not possible to provide remote employment opportunities that would allow you to work from another country.

To apply, please provide a detailed CV, a motivation letter and an outline of previous work or research experience along with the names of up to three referees (if available). Applications from female candidates are particularly encouraged. All other things being equal, candidates with disabilities will be given preference. Please submit your application via the following form by August, 31st: https://careers.ukp.informatik.tu-darmstadt.de/ukprecruitment.

Applications arriving after the deadline will still be considered if the position is not yet filled.